基于关联规则的灵活机器学习模块
Association Rule Based Flexible Machine Learning Module for Embedded System Platforms like Android
摘要:
在过去的几年里,智能手机的普及率有了巨大的增长。随着智能手机不断添加新功能以提高其实用性,其重要性在未来只会增加。将机器学习与移动计算相结合,可以使智能手机成为“智能”设备,这是迄今为止智能手机所没有的功能。此外,机器学习和上下文感知计算的组合可以使智能手机能够根据用户的环境和上下文主动衡量用户的需求。因此,可以向用户提供必要的服务。本文探讨了在Android平台上集成机器学习和上下文感知计算的方法和应用,为用户提供更高的实用性。为了实现这一点,我们定义了一个机器学习(ML)模块,该模块被纳入基本的Android架构。首先,我们概述了ML模块应该提供的两个主要功能。然后,我们介绍了三种架构,每种架构都在Android架构的不同级别上集成了ML模块。这些架构的优点和缺点进行了评估。最后,我们解释了几个应用程序,其中我们提出的系统可以被纳入,使他们的功能得到改善。
英文摘要:
The past few years have seen a tremendous growth in the popularity of smartphones. As newer features continue to be added to smartphones to increase their utility, their significance will only increase in future. Combining machine learning with mobile computing can enable smartphones to become 'intelligent' devices, a feature which is hitherto unseen in them. Also, the combination of machine learning and context aware computing can enable smartphones to gauge user's requirements proactively, depending upon their environment and context. Accordingly, necessary services can be provided to users. In this paper, we have explored the methods and applications of integrating machine learning and context aware computing on the Android platform, to provide higher utility to the users. To achieve this, we define a Machine Learning (ML) module which is incorporated in the basic Android architecture. Firstly, we have outlined two major functionalities that the ML module should provide. Then, we have presented three architectures, each of which incorporates the ML module at a different level in the Android architecture. The advantages and shortcomings of each of these architectures have been evaluated. Lastly, we have explained a few applications in which our proposed system can be incorporated such that their functionality is improved.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。